Explore our new Cybersecurity Terms Glossary, explained by our experts.

View the Glossary View the Glossary
PRIVIA

Cybersecurity Terms Glossary

What Is Privacy-Preserving Machine Learning?

Privacy-preserving machine learning is the set of techniques that let models be trained and used on sensitive data while limiting what anyone can learn about the individuals in it.

Illustration for the cybersecurity term Privacy-Preserving Machine Learning
Home / Blog / Cybersecurity Terms / What Is Privacy-Preserving Machine Learning?
17 November 2026

What Is Privacy-Preserving Machine Learning?

What Is Privacy-Preserving Machine Learning?

Privacy-preserving machine learning covers the techniques that allow a model to be built or used without exposing the data behind it. The need is practical rather than academic: a hospital trust wants to train on patient records, a bank wants to pool fraud signals with its peers, an insurer wants to analyse claims, and in each case the data is both essential and legally constrained. The techniques differ in what they protect against, which is the first thing to establish before choosing one.

Matching the Technique to the Threat

TechniqueProtects againstMain cost
Differential privacyInferring any individual from outputs or a trained modelAccuracy, especially for small subgroups
Federated learningCentralising raw data in one placeComplexity; updates still leak without added protection
Homomorphic encryptionThe compute provider seeing the data at allSevere performance overhead
Secure multi-party computationAny single party seeing the joint inputCommunication cost, protocol complexity
Trusted execution environmentsThe host operating system and operatorHardware trust assumptions, side-channel history
Synthetic dataDirect exposure of real recordsFidelity, and residual privacy risk

What Each One Actually Does

Differential privacy is the only approach offering a formal, measurable guarantee about individual exposure, applied in machine learning by clipping per-example gradients and adding calibrated noise during training. Federated learning avoids pooling data but exchanges model updates that can themselves leak, which is why it is normally combined with secure aggregation and differential privacy rather than used alone. Homomorphic encryption allows computation on encrypted values—the CKKS scheme supports the approximate arithmetic neural networks need—but remains orders of magnitude slower than plaintext computation, so it is used for narrow inference tasks rather than training. Secure multi-party computation splits a computation so that no participant sees the others’ inputs, which suits cross-institution analytics with a small number of parties. Trusted execution environments take a different approach entirely: the data is decrypted, but only inside hardware-isolated memory that the host cannot read, and modern GPU confidential computing has made this viable for model training at scale.

Combinations Rather Than Choices

Deployments rarely rest on one technique. A cross-hospital research programme might federate training, aggregate updates securely, apply differential privacy to bound individual contribution, and run the aggregator inside an enclave so the coordinating institution cannot inspect anything either. Each layer addresses a different adversary: a curious coordinator, a colluding participant, a compromised host, an external analyst reading published results. Specify the adversary first; it is the only way to tell whether a proposed architecture is proportionate or theatre.

The UK Regulatory Fit

The ICO treats these methods as privacy-enhancing technologies and addresses them in its guidance on anonymisation, pseudonymisation, and PETs. Three points matter for compliance teams. First, using a PET is strong evidence of data protection by design under Article 25 and of data minimisation, which is exactly what a Data Protection Impact Assessment should record. Second, it does not automatically take data outside UK GDPR—identifiability remains the test, assessed on the parameters you chose and what else an attacker could reasonably obtain. Third, the model produced may itself be personal data if individuals can be inferred from it, which is the membership inference question and a live one for erasure requests. In health and public sector research the UK has pushed secure data environments, where analysts work inside a controlled setting rather than receiving extracts, and these techniques complement that model rather than replacing it. The ICO and the US authorities also ran a joint PETs prize challenge, which produced useful public reference implementations for financial crime and health use cases.

Getting the Decision Right

Start with the question the model must answer and the specific legal constraint you face, then choose the lightest technique that satisfies both. Overengineering is common and expensive; so is applying a technique with parameters so weak that the guarantee is nominal. Document the threat model, the parameters, and the residual risk, because that documentation is the deliverable a regulator or a partner institution will ask for. At Privia Security, our Cyber Risk Analysis Services assess whether a proposed privacy architecture addresses the adversary you actually have.

FAQ – Frequently Asked Questions

Which Technique Should We Start With?

Identify the adversary first. If you are worried about what a published model or result reveals about individuals, differential privacy is the direct answer. If the obstacle is that data cannot leave each institution, federated learning addresses that. If the concern is the cloud provider seeing your data, confidential computing fits. Choosing before defining the threat usually produces an expensive architecture that misses the real risk.

Is Homomorphic Encryption Practical Yet?

For narrow inference tasks on small models, yes; for training large models, not at present. The performance overhead remains several orders of magnitude, and schemes supporting the operations neural networks need are the slowest. Most production systems that need a strong confidentiality guarantee use trusted execution environments instead, accepting a hardware trust assumption in exchange for usable performance.

Do These Techniques Remove UK GDPR Obligations?

No, though they can change which obligations bite. The ICO assesses identifiability on the facts, so a strong, well-parameterised approach may support an anonymisation argument while a weak one will not. Even where data remains in scope, applying a PET is persuasive evidence of data protection by design and of appropriate security measures under Article 32.

Can We Apply Them to an Existing Model?

Mostly not. Differential privacy and federated learning determine how a model is trained, so they require retraining. Confidential computing and access controls can be applied to an existing model’s serving environment, and output minimisation and rate limiting reduce what an attacker can learn from it. For an existing high-risk model, test it for memorisation and plan a retrain if the results are poor.

Related Terms

You May Be Interested In These

What Is a Bootkit?

Cybersecurity Terms

17 November 2026

What Is a Bootkit?

Read More
What Are Disk Imaging and Write Blockers?

Cybersecurity Terms

17 November 2026

What Are Disk Imaging and Write Blockers?

Read More
What Is Port Forwarding?

Cybersecurity Terms

17 November 2026

What Is Port Forwarding?

Read More